The backbone of modern engineering: CAD software. From ingenious consumer products to finely crafted machine parts and building designs, everything starts with a precise digital drawing. But anyone who’s ever used CAD software knows how steep the learning curve can be. Mastering all the commands and workflows can take months—sometimes years.
The golden rule: software should truly unburden the designer. Hence the "A" in CAD: Aided. Many CAD vendors aim for maximum automation, where the user hardly has to design at all. Press a button, and the 3D model appears.
Full automation leads to black-box applications: users see the end result, but not the how and why. This breeds distrust and causes them to manually check and adjust changes.
At Bricsys, with an Innovation Mandate supported by VLAIO (Flemish Institute for Innovation and Entrepreneurship), we’re consciously taking a different path: that of guided automation, where smart suggestions go hand in hand with full user control.
This article is part of our ongoing series on AI in BricsCAD. If you missed the introduction, you can read it here: What Does “AI-Driven” Really Mean in CAD?
Complexity versus usability
While CAD software offers enormous power and limitless possibilities, it can also be a dense jungle of commands and options. Designers lose valuable time deciphering automated routines when what they really need is speed and reliability.
The challenge is to find a middle ground. Guided automation means the software makes smart suggestions, but the user can continue to make choices step by step based on the information provided. This way, they retain control and still save time.
This approach requires both powerful algorithms to generate relevant proposals and a user interface that clearly explains why certain parameters are chosen and how changes are reflected in the design. For that, Bricsys turned to the Digital Future Lab department at UHasselt, where Prof. Dr. Raf Ramakers leads the Computational Design and Fabrication research group.
Intelligibility: the finesse of human–machine interaction
The Digital Future Lab focuses on improving how humans and machines interact. “Intelligent systems are great,” said Ramakers, “but they also need to be understandable and controllable. We achieve that by making hidden processes visible. In our jargon, this is called intelligibility.
“Your smart system shouldn’t just generate solutions; it should also be able to explain them clearly, in words or visually. Intelligible software provides insight into the decisions it makes, why it proposes certain options, and the advantages and disadvantages of each solution. This not only makes CAD more efficient, but it also shortens the learning curve because users truly understand what’s happening.”
He emphasizes that this approach doesn’t mean simplifying CAD by stripping away features: “We retain the power of professional CAD software, but we make its complexity manageable by providing context and feedback step by step. The engineer keeps control and saves time.”
From theory to practice
During his PhD research, Dr. Tom Veuskens, as part of Ramakers’s lab at UHasselt, conducted experiments with a prototype for parametric 3D modeling. “We discovered that users don’t trust fully automatic solutions,” he explains. “They want to understand exactly how everything works and verify that the results are correct. My conclusion? The time saved through a fully automated system was often negated by the constant checking and adjustments.”
Veuskens is now continuing his research at and for Bricsys through a VLAIO Innovation Mandate designed to connect academic research with industry. Such a mandate gives Bricsys the means to engage Tom as a top talent and ensures the continuous involvement of our internal product owners, support data analysts and R&D team.
“I'm getting a taste of what it means to work in a more applied, less abstract way, and to justify everything before you do anything,” said Veuskens. “It's a little less of a playground where you can freely experiment with cool ideas. But it's still fun, of course, to integrate your prototypes directly into BricsCAD as a researcher and developer. I like that idea of having it both ways."
Looking ahead
The project is still in the research phase, but the goal is clear: to integrate a feature in BricsCAD that supports parameterization and dimensioning through clear, guided interaction. This should reduce repetitive drawing work, accelerate the design-to-production process, and make the software more accessible to SMEs without demanding heavy training investments.
For the university, this collaboration is equally valuable. “Such intensive, practice-oriented collaboration provides new insights into the usefulness and usability of our intelligibility solutions,” said Ramakers. “We’ve demonstrated the basic principles and now we can integrate them within a much more complex system.”
At Bricsys, we see this as another step toward CAD software that doesn’t replace human creativity but enhances it.
Have you registered for the BricsCAD V26 launch event? Don't miss your chance to join as as we unveil 60+ new features.
